A statistical framework for the description of stochastic resonance phenomena (Q2722400)

This paper shows how statistical distance measures known as \(\phi\)-divergences can provide a general framework for stochastic resonance type phenomena. This concept is then generalized to mean local maximization of \(\phi\)-divergences rather than signal-to-noise ratio measures.NEWLINENEWLINENEWLINEThis provides a general definition of stochastic resonance being applicable to virtually any kind of systems or noise and it establishes a firm link to various bounds in information theory or statistics. Further, it is shown that signal-to-noise ratio-based measures are upper bounded by a certain \(\phi\)-divergence, and information conserving (invertible) systems preserve \(\phi\)-divergences.NEWLINENEWLINEFor the entire collection see [Zbl 0958.00031].